1 Effects of forest management on bryophyte species richness in Central European forests
2 Jörg Müller1,2*, Steffen Boch3,4*, Daniel Prati3, Stephanie A. Socher3,6, Ulf Pommer1,7,
3 Dominik Hessenmöller5,8, Peter Schall9, Ernst Detlef Schulze5, Markus Fischer3
4
5 1Institute for Biochemistry and Biology, University of Potsdam, Maulbeerallee 1, 14469
6 Potsdam, Germany
7 2Department of Nature Conservation, Heinz Sielmann Foundation, Unter den Kiefern 9,
8 14641 Wustermark, Germany
9 3Institute of Plant Sciences and Botanical Garden, University of Bern, Altenbergrain 21, CH-
10 3013 Bern, Switzerland
11 4Swiss Federal Research Institute WSL, Zürcherstrasse 111, CH-8903 Birmensdorf,
12 Switzerland
13 5Max-Planck-Institute for Biogeochemistry, Hans-Knöll-Strasse 10, 07745 Jena, Germany
14 6Department of Biosciences, University of Salzburg, 5020 Salzburg, Austria
15 7Senckenberg Gesellschaft für Naturforschung, Biosphere reserve Schorfheide-Chorin, Hoher
16 Steinweg 5–6, 16278 Angermünde, Germany
17 8Forstamt Schmalkalden, Thueringen Forst, Schlossberg 11, 98574 Schmalkalden, Germany
18 9Department of Silviculture and Forest Ecology of the Temperate Zones, University of
19 Göttingen, Büsgenweg 1, 37077 Göttingen, Germany
20
21 *These authors contributed equally to this work.
22
23 Correspondig author: Steffen Boch (steffen.boch@wsl.ch)
24
25 Keywords: Beech forests, conifer plantations, cryptogams, ecological guilds, forest
26 management, temperate forests, selection vs. age-class forests; unmanaged vs. managed
This document is the accepted manuscript version of the following article:
Müller, J., Boch, S., Prati, D., Socher, S. A., Pommer, U., Hessenmöller, D., … Fischer, M. (2019).
Effects of forest management on bryophyte species richness in Central European forests. Forest Ecology and Management, 432, 850-859. https://doi.org/10.1016/j.foreco.2018.10.019
This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/
27 forests, woodland indicator species
28 Abstract
29 We studied the effect of three major forest management types (unmanaged beech, selection
30 beech, and age class forests) and stand variables (SMId, soil pH, proportion of conifers, litter
31 cover, deadwood cover, rock cover and cumulative cover of woody trees and shrubs) on
32 bryophyte species richness in 1050 forest plots in three regions in Germany. In addition, we
33 analysed the species richness of four ecological guilds of bryophytes according to their
34 colonized substrates (deadwood, rock, soil, bark) and the number of woodland indicator
35 bryophyte species. Beech selection forests turned out to be the most species rich management
36 type, whereas unmanaged beech forests revealed even lower numbers than age-class forests.
37 Increasing conifer proportion increased bryophyte species richness but not the number of
38 woodland indicator bryophyte species. The richness of the four ecological guilds mainly
39 responded to the abundance of their respective substrate. We conclude that the permanent
40 availability of suitable substrates is most important for bryophyte species richness in forests,
41 which is not stringently linked to management type. Therefore, managed age-class forests and
42 selection forests may even exceed unmanaged forests in bryophyte species richness due to
43 higher substrate supply and therefore represent important habitats for bryophytes. Typical
44 woodland indicator bryophytes and their species richness were negatively affected by SMId
45 (management intensity) and therefore better indicate forest integrity than the species richness
46 of all bryophytes. Nature conservation efforts should focus on the reduction of management
47 intensity. Moreover, maintaining and increasing a variability of substrates and habitats, such
48 as coarse woody debris, increasing structural heterogeneity by retaining patches with groups
49 of old, mature to over-mature trees in managed forests, maintaining forest climate conditions
50 by silvicultural methods that assure stand continuity, e.g. by selection cutting rather than clear
51 cutting and shelterwood logging might promote bryophyte diversity.
53 1. Introduction
54 One of the most significant landscape elements in Central Europe are forests. In Germany,
55 approximately 31% of the surface is covered by mainly managed forests. Natural or
56 unmanaged forests exist only in fragments, comprising less than 3% of total forest area
57 (BMELV, 2014). Ongoing efforts to increase the proportion of green energy focus on
58 additional harvest of wood as well-meant strategy to counteract climate change (Schulze et
59 al., 2012) act as additional threat for forest ecosystems because often these short-rotation
60 plantations are mainly composed of fast-growing tree species which are often untypical for
61 the sites where they are cultivated. Therefore, ecologically sustainable and conservation-
62 compatible practices in managed forests, such as integrative multifunctional forest
63 management programs (e.g. Borrass et al., 2017), as well as the abandonment of formerly
64 managed forests gain importance for the conservation of biodiversity in the European
65 landscape. Moreover, Ammer et al. (in press) highlighted the importance of understanding the
66 relationships between different components of management, forest structure and biodiversity.
67 This demands the solid investigation of environmental, structural and forest management
68 variables to test their effects on biodiversity. Usually, studies investigating management
69 effects on biodiversity comprise species groups such as mammals, birds, beetles or higher
70 plants whereas equally important forest organisms such as cryptogam plants are strongly
71 underrepresented (but see Paillet et al., 2010; Boch et al., 2013a, b).
72 Among cryptogams, bryophytes constitute an important and permanent component of the
73 forest flora and diversity. They colonize various substrates, which are unsuitable for vascular
74 plants, because of low light intensity or low nutrient level, such as deadwood, bark, rocks, and
75 open soil (Smith, 1982; Bates, 1992). They provide shelter habitats, food, and nest material
76 for many animals (Gerson, 1982; Boch et al., 2013c). Due to their high abundance in forests
77 they are considerable primary producers of biomass, conduct photosynthesis around the year,
78 and are therefore influencing the carbon and nutrient cycles (e.g. Turetsky, 2003).
79 In forests, different ecological guilds of bryophytes can be distinguished by the substrate on
80 which they are growing, including terricolous, lignicolous, corticolous and saxicolous species
81 that occur on soil, deadwood, bark of living trees and shrubs, or rocks, respectively. As
82 diversity and quality of these substrates is affected by forest management, bryophytes are
83 suitable indicators for the effect of management on forest conditions (Rose, 1992). Especially
84 typical woodland bryophytes, which are strictly depending on forest conditions (Rose, 1999;
85 Nordén et al., 2007; Preussing et al., 2011; Schmidt et al., 2011; Kriebitzsch et al., 2013,
86 Mölder et al., 2015), are valuable indicator organisms to estimate the naturalness and integrity
87 of forest stands (Frego, 2007).
88 Some studies dealt with forest management effects on bryophytes and demonstrated their
89 sensitivity to management practices. However, comparative studies on bryophyte diversity in
90 forest ecosystems were mostly restricted to boreal regions and only compared unmanaged
91 with managed forests (e.g. Andersson and Hytteborn, 1991; Frisvoll and Prestø, 1997; Ross-
92 Davis and Frego, 2002; Nelson and Halpern, 2005; Vellak and Ingerpuu, 2005; reviewed in
93 Paillet et al., 2010). The studies from temperate European forests were restricted to only one
94 study region and one ecological guild such as epiphytes (Friedel et al., 2006; Mežaka and
95 Znotiņa, 2006; Bardat and Aubert, 2007; Fritz et al., 2008, 2009a, 2009b; Kiraly and Ódor,
96 2010) or epixylic species (Andersson and Hytteborn, 1991; Ódor et al., 2006) or incorporated
97 only one forest type (Márialigeti et al., 2009; Horvat et al., 2017; Schall et al., 2018). Thus, a
98 comprehensive assessment how different groups of bryophytes respond to a range of
99 management regimes across different regions is still lacking.
100 In addition, the relationship between the diversities of bryophytes and vascular plants has only
101 rarely been studied in forests (but see Ingerpuu et al., 2003), and it is unclear whether
102 management affects both groups similarly. In grasslands, bryophyte diversity is a very good
103 indicator not only for the total vascular plant diversity but also for the diversities of many
104 individual plant and animal taxa. In fact, bryophyte diversity showed the strongest relation
105 with of the overall diversity of grasslands (Manning et al., 2015 referring to multidiversity
106 sensu Allan et al., 2014). In forests, the richness of terricolous bryophyte and herb species
107 seems to be positively related as well (Márialigeti et al., 2009). However, it remained open
108 whether particular forest conditions affect the presence of typical woodland indicator species
109 of both bryophytes and vascular plants in the same way.
110 We therefore conducted a large-scale investigation on the relationship between forest
111 management and bryophyte species richness across a broad range of management types in
112 three regions in Germany, allowing to generalize our findings for temperate, lowland to
113 montane forests in Central Europe. We included bryophytes of all ecological guilds and
114 analysed management responses of each ecological guild, of the group of woodland indicator
115 bryophyte species and of all bryophyte species together. Furthermore, to test the accordance
116 of management responses on the vegetation, we studied the relationship between species
117 richness of bryophytes and herbaceous vascular plants on the forest floor.
118 Our main questions are:
119 1) How does the species richness of all bryophytes and of woodland indicator bryophyte
120 species respond to forest management?
121 2) How does bryophyte species richness respond to variation in environmental variables?
122 3) How is the richness of bryophytes and herbaceous vascular plants related?
123
124 2. Methods:
125 2.1 Study system
126 The investigations were carried out within the large-scale and long-term ‘Biodiversity
127 Exploratories’ project, which provides an advantageous framework to analyse land-use effects
128 on different species groups including bryophytes. We studied bryophytes in 1050 forest plots
129 in three regions of Germany (Fig. A.1) which differ in geological and climatic conditions
130 (Fischer et al., 2010). Moreover, the historic forest management likely differed between the
131 regions and still might affect the current forest attributes on the landscape level (cf. Wäldchen
132 et al., 2011).
133 The most northern region is the UNESCO-Biosphere reserve Schorfheide-Chorin, located in
134 the northeast of Germany. It is a post glacial moraine composed landscape, on mainly sandy,
135 acidic soils with variable proportions of loam. The climate has a slightly sub-continental
136 character and has the lowest mean annual precipitation of the three study regions with 520 to
137 600 mm.
138 The second region in the middle of Germany, the Hainich-Dün, comprises the National Park
139 Hainich and its surrounding areas in a mid-elevation mountain range, with a mean annual
140 precipitation of 750–800 mm. This region is characterized by shell limestone and Loess soils
141 bearing one of the largest unfragmented European Beech forests of Germany.
142 The third region is situated in the southwest of Germany in the Biosphere area Schwäbische
143 Alb. This area is part of the Swabian Jura, a middle mountain range formed by calcareous
144 bedrock. Its climate has a montane character with relatively high mean annual precipitation of
145 935–965 mm. The forests in this region are more fragmented than the ones in the other study
146 regions.
147 For the Biodiversity Exploratories project approx. 500 plots were established in each region.
148 The plots were selected from the intersection points of a 100 m × 100 m grid distributed
149 across forested area, after discarding forest edges, i.e. plots fully or partially overlapping with
150 settlements, grasslands, agricultural fields, water bodies, and rare azonal forest types such as
151 floodplain forests, as well as plots intersected by roads, with the aim to represent the most
152 common forest and management types in the particular region (see Fischer et al., 2010). They
153 represent the whole regional range of forest management types, comprising (i) unmanaged
154 and uneven-aged forests dominated by European beech (Fagus sylvatica); (ii) selection forests
155 which were also mainly dominated by European beech and which is a type of continuous-
156 cover forest where individual trees or small groups of trees are harvested; and (iii) age-class
157 forests which are dominated by European beech in all three regions, or Norway spruce (Picea
158 abies) in Hainich-Dün and Schwäbische Alb or Scots pine (Pinus sylvestris) in Schorfheide-
159 Chorin. The age-class forests comprised different developmental stages of even-aged structure
160 due to harvests at 80 to 120 year intervals, either by conventional clear cuts or shelterwood
161 logging. For this study, we randomly selected a representative subset of 1050 out of 1535
162 plots: 441 in Schorfheide-Chorin, 271 in Hainich-Dün, and 338 in the Schwäbische Alb. The
163 plots had a size of 20 m × 20 m and were restricted to forest interiors.
164
165 2.2 Management and environmental variables
166 In addition to the three main management types of unmanaged, selection, and age-class
167 forests, we recorded height and diameter at breast height (DBH) of each tree in a circular
168 forest inventory area of 500 m2, which was concentric with our plots and calculated the basal
169 area (BA) of each plot. We then quantified forest management intensity (SMId sensu Schall
170 and Ammer, 2013) by relating the basal area to their carrying capacity (BAcc). Forest
171 management intensity thus measures the deviation of the actual stocking from a fully stocked
172 mature forest (= 1 – BA × BAcc-1), due to young stand age, harvests and thinnings. Basal area
173 carrying capacity was quantified species specifically using the 95% quantile of observed
174 values for beech dominated forests as reference (45 m2 ha-1). Subsequently the carrying
175 capacity of other tree species were estimated relative to beech based on yield tables (oak and
176 other broadleaves: 36 m2 ha-1; pine and larch: 51 m2 ha-1; spruce and fir: 63 m2 ha-1; Douglas-
177 fir: 69 m2 ha-1). In mixed stands, forest management intensity was quantified relative to the
178 current tree species composition (= 1 – Σ BAi × BAcc,i-1, with i representing tree species). As
179 this measure does not account for different tree species, it was not possible to use it
180 independently from our management categories.
181 For age-class forests, we used the proportion of conifers based on canopy cover estimations as
182 a continuous measure (see below) instead of using coniferous, deciduous and mixed forests as
183 categories. Furthermore, we estimated the cover of litter, bare ground, deadwood, and rocks
184 for a measure of substrate availability for different guilds of bryophytes.
185 Close to the center of each plot, a soil sample was taken from the uppermost soil horizon (A,
186 H or E horizon), which comprised approximately the uppermost 10 cm of the soil core. The
187 samples were air dried and sieved to <2 mm. The soil pH was determined with a glass
188 electrode in the supernatant of a soil suspension using a 1:2.5 mixture of soil and 0.01 M
189 CaCl2.
190
191 2.3 Vegetation data
192 Between April 2007 and September 2008, we sampled bryophyte species in all plots and
193 estimated their percentage cover in each of four ecological guilds, according to their
194 colonized substrates on the plot. The four guilds comprised terricolous bryophytes on the
195 forest floor, lignicolous bryophytes on deadwood, saxicolous bryophytes on rocks or stones,
196 and corticolous bryophytes (epiphytes) on the bark of living trees and shrubs including
197 species growing up to a height of 2 m. This method probably underestimates the overall
198 bryophyte species richness because we could not assess species restricted to tree crowns.
199 However, it has been shown that sampling many tree individuals in a stand, i.e. the level of a
200 plot, keeps the number of overlooked bryophyte species relatively low (Boch et al., 2013d;
201 Kiebacher et al., 2016).
202 To investigate the relationship of bryophytes to forest vegetation, we recorded all vascular
203 plant species of the forest floor and their percentage cover, separately for herbaceous species
204 (including ferns and seedlings of woody species), shrubs (woody species from >0 to 5 m
205 height), first tree layer (tree height between 5 and 10 m), and second tree layer (trees higher
206 than 10 m) in spring and summer. We then calculated the cumulative cover of the two tree
207 layers and the shrub layer (called ‘cumulative cover of woody species’ henceforth) as a proxy
208 for light availability in the stands (cf. Ewald et al., 2011). Nomenclature of vascular plants
209 followed Wisskirchen and Häupler (1998).
210 Afterwards, we classified all recorded bryophyte and herb species into categories according to
211 their typical habitat requirements with regard to forest affinity and forest dependency, i.e.
212 typical woodland species, according to Preussing et al. (2011) for bryophytes and Schmidt et
213 al. (2011) for vascular plants. For each plot, we counted the numbers and calculated the
214 proportions of typical woodland indicator species of bryophytes and herbs, defined as those
215 which mainly occur in closed forests (= forest specialist species; category 1.1 sensu Preussing
216 et al., 2011; Schmidt et al., 2011). The numbers of typical woodland indicator bryophyte
217 species were used to test whether these species are better indicators for forest integrity
218 (Preussing et al., 2011).
219
220 2.4 Data analyses
221 To examine the relationship between forest management and bryophyte species richness we
222 conducted three separate analyses of co-variance, because of the non-orthogonality of
223 management types and dominant tree species. In the first model, we analysed differences
224 between managed and unmanaged forests and included all plots. In the second model, we
225 tested different management types (unmanaged vs. selection vs. age-class) but considered
226 only deciduous stands. As selection forests were restricted to deciduous forests of the
227 Schwäbische Alb and Hainich-Dün, we excluded the Schorfheide-Chorin region from this
228 analysis. In the third model, we tested the effect of the proportion of coniferous trees within
229 the subset of age-class forests. As we were analysing count data, we chose GLM models with
230 Poisson errors. Sequential Chi-square tests were used to analyse the significance of deviance
231 changes among our predictors associated with factors added to the model in the sequence
232 shown in Tabs. 1–3. In all three models, region and the co-variates region, soil pH, rock
233 cover, deadwood cover, litter cover and cumulative cover of woody species (trees and shrubs)
234 were fitted first. This way, we corrected the effect of forest management for regional
235 differences and differences in habitat variables and substrate availability. No model selection
236 was performed, but substrate availability was included only for corresponding groups of
237 bryophytes. In addition, we included interactions between co-variates and region.
238 Furthermore, we included the SMId as a proxy for management intensity and its interaction
239 with management and region, as well as the region-by-management interaction into the
240 model. We analysed species richness of all bryophytes, as well as woodland indicator
241 bryophyte species and the four ecological guilds, separately. In addition, we calculated linear
242 regressions to test the relationship between species richness of herbaceous plants and
243 terricolous bryophytes as well as between woodland indicating bryophytes and herbs. Data
244 were analysed using R, Version 3.3.3 (RCORE TEAM, 2017) and the nlme package (Pinheiro
245 et al., 2017).
246 247
248 3. Results
249 3.1 Overall and regional species richness
250 Overall, we found 186 different bryophyte species (167 mosses, 19 liverworts). Most species
251 occurred in Schwäbische Alb (131 mosses, 14 liverworts), followed by Hainich-Dün (105
252 mosses, 10 liverworts), and Schorfheide-Chorin (72 mosses, 12 liverworts). The bryophyte
253 species numbers per plot ranged from 2 to 45 species (overall mean 12.7 ± 0.2 SE species per
254 400 m2). Mean values differed significantly among the regions, whereas Schwäbische Alb had
255 highest species richness and Schorfheide-Chorin the lowest (Tabs. 1–3, A.1, Fig. A.2). This
256 pattern was mainly caused by an evident higher number of corticolous and saxicolous species
257 in Schwäbische Alb compared with the other two regions (Tab. A.1). Regarding mean
258 terricolous and lignicolous species numbers, we found less pronounced differences among the
259 regions and in case of lignicolous species even no regional differences within age-class forests
260 (Tabs. 1–3, A.1).
261
262 3.2 Management induced differences in species richness
263 In general, bryophyte species richness was higher in managed than in unmanaged forests
264 (Tabs. 1, A.1, Fig. 1a). This difference was mainly because of higher numbers of lignicolous
265 and terricolous species in managed than in unmanaged forests (Tab. A.1, Fig. 1a). However,
266 within the regions we observed this pattern only in Hainich-Dün and Schorfheide-Chorin. In
267 the Schwäbische Alb we found the opposite result due to high numbers of saxicolous and
268 corticolous species, which compensated the low numbers of lignicolous and terricolous
269 bryophyte species in unmanaged European beech forests (Tab. A.1). Increasing management
270 intensity (SMId) decreased the total species richness, mainly because of decreasing numbers
271 of corticolous and saxicoulous species. This pattern was consistent among management type,
272 as indicated by the non-significant SMId × management interaction (Tab. 1)
273 Among deciduous forests, the selection forests harboured the highest number of total,
274 corticolous and saxicolous species (Tabs. 2, A.1, Fig. 1b). Within age-class forests we
275 observed a positive relation between the proportion of conifer trees in the canopy and the total
276 species richness, mainly due to strongly increasing numbers of lignicolous and terricolous
277 bryophytes, while saxicolous and corticolous species numbers decreased (Fig. 2).
278
279 3.3 Environmental variables and bryophyte species richness
280 The richness of terricolous bryophyte species was decreasing with increasing soil pH values
281 (Tab. 1). Increasing substrate availability of rocks or deadwood had strong benefiting effects
282 on the number of saxicolous or lignicolous species, which also resulted in increasing total
283 species numbers (Tab. 1). Increasing cover of litter decreased terricolous species richness and
284 therefore also total species richness, which was also indicated by the significant interaction
285 between conifer proportion and litter cover (Tab. 3). Overall, increasing cumulative cover of
286 woody species (shrubs and trees) had positive effects on corticolous and negative effects on
287 lignicolous and terricolous species. Within deciduous and age-class forests it additionally
288 negatively affected the richness of saxicolous species (Tabs. 2, 3). Similarly to the direct
289 management effects, the patterns between management-related variables and bryophytes were
290 not consistent among regions and management types as indicated by significant interactions
291 (Tabs. 1–3).
292
293 3.4 Woodland indicator bryophyte species
294 On average the number of woodland indicator bryophyte species was 2.9 (± 0.1 SE) per plot.
295 This corresponds to a mean proportion of 22.8% of the total species richness. Generally,
296 woodland indicator bryophyte species were positively correlated to total bryophyte species
297 richness (Fig. A.3). Therefore, numbers of woodland indicator bryophyte species revealed
298 similar patterns of region and management effects compared to total species richness (Figs.
299 3a, 3b), with the exception of conifer proportion in the canopy that had no benefiting effects
300 on the richness of woodland indicator bryophyte species within age-class stands, although it
301 increased the total species richness significantly (Fig. 2). Corresponding to patterns of total
302 species richness, we observed also benefiting effects of increasing rock cover, as well as
303 negative effects of management intensity on the richness of woodland indicator bryophyte
304 species across and within the studied forest types of deciduous and age-class stands. However,
305 again these patterns and their significance differed among regions and management types as
306 indicated by significant interactions (Tabs. 1–3).
307
308 3.5 Relation of herb and bryophyte species richness
309 Generally, the richness of terricolous bryophyte and herbaceous vascular plant species was
310 positively related both across and within regions. In Schwäbische Alb we observed the
311 closest, in Hainich-Dün the weakest correlation between the richness of these plant taxa (Fig.
312 3a). Overall, the number of woodland indicator bryophyte species was positively related to the
313 number of woodland indicator herb species (Fig. 3b). However, the correlation was weak and
314 no longer significant when analysed separately for the three regions.
315
316 4. Discussion
317 4.1 Regional effects
318 Overall, bryophytes proved to be important parts of forest vegetation in all study regions.
319 Regional differences in species richness are probably caused by considerable climatic (most
320 importantly precipitation) and geological differences (e.g. landscape texture, soil types)
321 leading to differing substrate availability and suitability of forest habitats for bryophytes
322 (Vanderpoorten and Engels, 2002; Bates et al., 2004). Furthermore, the regions vary in air
323 quality with regard to former and current atmospheric pollution (see Boch et al., 2013a) and
324 eutrophication which are considered to have strong impacts on bryophytes (e.g. Friedel and
325 Müller, 2004; Mitchell et al., 2005; Davies et al., 2007). Finally, the historic forest use over
326 the last centuries differed among the regions and continued to affect the current forest
327 attributes on the landscape level (Wäldchen et al., 2011). Given all these fundamental
328 differences among the study regions, this highlights the importance to investigate more than
329 one region to detect general patterns between forest management and bryophyte vegetation.
330
331 4.2 Effects of forest management and environmental variables on bryophytes
332 Generally, the studied managed forests had more species than the unmanaged ones. Especially
333 age-class forests with high proportions of conifers revealed high species densities due to high
334 numbers of lignicolous and terricolous species. Lignicolous species generally benefited from
335 the abundance of suitable deadwood substrates (e.g. Humphrey et al., 2002), which was larger
336 in the studied coniferous forests probably due to retention of logging debris and stumps
337 therein (see also Müller et al., 2015). Furthermore, special attributes of spruce or pine
338 deadwood such as their resin content, high acidity, low nutrient, high moisture availability
339 and slow decomposition (e.g. Fengel and Wegener, 1984; McAlister, 1995; Kahl et al., 2017)
340 are responsible for the colonization of specialist bryophyte species adapted to acidic and
341 nutrient poor conditions of deadwood, resulting in highly diverse bryophyte communities on
342 coniferous stumps and logs. In contrast, beech deadwood has a moderate pH and lacks resin
343 (Fengel and Wegener, 1984) which results often in high abundances of only few, highly
344 competitive and opportunistic moss species.
345 Interestingly, the richness of terricolous bryophyte species was decreasing with increasing soil
346 pH values, which is in contrast to many other studies reporting a positive bryophyte diversity-
347 soil pH relationship (e.g. Tyler, T. 2005; Tyler et al., 2018) and contradicts the general
348 opinion that soil acidification might be responsible for the decline of bryophyte species
349 richness (e.g. Delgado and Ederra, 2013). However, this finding might be masked by a tree
350 species effect, as mean soil pH is on average lower in our coniferous forests plots, than to one
351 of the deciduous forest plots (3.9 ± 0.06 SE in coniferous vs. 4.8 ±0.05 in deciduous forest
352 plots): the richness of terricolous bryophyte species in conifer forests likely profited from
353 higher light availability on the ground due to a less closed canopy (Tinya et al., 2009) and
354 reduced litterfall compared with deciduous forests. These conditions facilitate the occurrence
355 of thick mats of terricolous bryophyte species composed of feather mosses in conifer forests,
356 which are rather typical for montane or boreal forests (Ross-Davis and Frego, 2002; Nelson
357 and Halpern, 2005; Vellak and Ingerpuu, 2005). In contrast, in beech forests the conditions
358 are rather unfavorable for terricolous bryophytes (Márialigeti et al., 2009). Their presence
359 relies on bare soil patches (e.g. at slopes) or occasional soil disturbances, which remove at
360 least temporarily the thick litter layers that hinders bryophyte growth (Dzwonko and
361 Gawroński, 2002). Such soil disturbances can be created by animals (wild boar), wind throw,
362 or forest management practices such as logging trails. The latter ones lead to additional
363 positive effects of management on microhabitat availability for bryophytes in managed
364 forests, and this may explain the high bryophyte diversity found in selection forests because
365 this type of forest management is characterized by repeated disturbances. However, we should
366 keep in mind that other taxa might depend on structures and processes of old growth forests
367 and low levels of anthropogenic disturbance and might be negatively affected by these
368 frequent disturbances (Bauhus et al., 2009). In contrast, bryophytes are potentially prone to
369 destruction by clear cuttings (e.g. Fenton et al., 2003; Nelson and Halpern, 2005), and
370 consequently, age-class forests may fail to provide any reliable long-term refuge for forest
371 species.
372 Low numbers of corticolous species observed in conifer dominated forests can be mainly
373 attributed to ‘epiphyte-repellent’ conditions of the stems of coniferous trees (Barkman, 1958;
374 Fengel and Wegener, 1984; Mežaka and Znotiņa, 2006; Kiraly and Ódor ,2010). However,
375 species-rich epiphyte communities on broad-leafed shrubs or little trees of European rowan
376 (Sorbus aucuparia) and elder (Sambucus nigra and S. racemosa) that grow in the understorey
377 of coniferous forests, can compensate for the lack of epiphytes on pine and spruce to a certain
378 amount and can provide suitable epiphyte habitats despite low density of broad-leafed trees
379 (Hazell et al., 1998).
380 Selection forest was the most bryophyte species-rich management type among all deciduous
381 forests. This forest type is characterized by uneven-aged structure of trees due to long-term,
382 continuous forest utilization by single tree harvest and continuous self-regeneration of beech.
383 These near-natural forest structures accompanied with forest-typical microclimate conditions
384 favor species-rich corticolous communities in contrast to homogenous structures in age-class
385 forests (Bardat and Aubert, 2007; Kiraly and Ódor, 2010; Boch et al., 2013a). In addition,
386 bryophytes in selection and age-class forests might profit from land-use induced habitat
387 availability, including retention of medium sized deadwood and stumps, and soil disturbances
388 on logging trails, benefiting lignicolous and terricolous species.
389 The number of bryophytes in unmanaged forests was surprisingly low in our study,
390 contradicting the findings of a generally negative effect of forest management on bryophyte
391 diversity reported by Horvat et al. (2017), who studied bryophyte diversity among differently
392 managed silver fir-beech forests in the western Pyrenees including stands which were
393 abandoned since >40 years, Kaufmann et al. (2017) who compared the bryophyte diversity of
394 primeval and production beech forests in the western Carpathian Mountains, and the ones
395 reviewed in Paillet et al. (2010). On the one hand, our results can be attributed to negative
396 long-term effects of land-use history and the slow regeneration of old-growth forests with
397 their typical attributes comprising a multilayered forest structure with a wide range of tree
398 ages, site-adapted mixed tree species compositions and large amounts of deadwood of
399 different type and decay stage (Standovar et al., 2006). Whereas some forest sites in
400 Schwäbische Alb and Hainich-Dün have been abandoned for more than 70 years, most
401 unmanaged forests have been abandoned only since two decades (Huss and Butler-Manning,
402 2006). Therefore, in some of the recently abandoned forests typical attributes of real natural
403 forests still lack. Additionally, due to decelerated re-colonization of slow dispersing
404 bryophytes to sites of enhanced forest attributes the bryophyte vegetation of unmanaged
405 forests might still suffer from legacy effects of former management impacts (Aude and
406 Lawesson, 1998; Snäll et al., 2004; Fritz et al., 2008). On the other hand, the negative
407 management effects on bryophytes reported by Paillet et al. (2010) stem mostly from boreal
408 regions (twelve out of fourteen studies analysed), and there is an urgent need to include more
409 temperate forest before we can draw general conclusions on management effects on
410 bryophytes.
411 Forest management intensity (SMId), which is negatively related to stand volume (in our case
412 R2 = 0.7239; p < 0.0001), was decreasing total bryophyte species richness, mainly because of
413 a general decrease in the richness of corticolous and saxicolous species. This reflects the
414 preference of these species for old stands and trees, providing more suitable microhabitats
415 than young ones (Gustafsson et al., 1992; Fritz et al., 2009a, 2009b; Boch et al., 2013a, d).
416 Furthermore, due to high tree age, more species had the opportunity to colonize their bark.
417 Stand-age is related to canopy density and multi-layered forest structures which probably
418 result in more forest-typical conditions of microclimate, comprising less variation in
419 temperature and generally higher air humidity, conditions considered to be favorable for
420 epiphytic species in general (Frahm, 2003; Bardat and Aubert, 2007), and forest specialists in
421 particular (McGee and Kimmerer, 2002).
422 In addition to forest management and site attributes, further factors likely determine
423 bryophyte species richness on the plot level, which we might have excluded from our study
424 because of the sampling design. For example the presence of suitable microhabitats and
425 substrate diversity on very small scale (Mills and MacDonald, 2005), which is not per se
426 related to forest management types and age-structure, also can provide important micro-
427 hotspots such as large stumps, steep ridges, or creeks even in unnatural tree plantations.
428 Furthermore, local species assemblages, representing only a proportion of the forests’ species
429 pool, fluctuate both in space and time (Zobel, 1997), which results in the deviation between
430 the potential and the actually observed bryophyte vegetation on particular sites. Finally,
431 management affects the bryophyte vegetation not only by altering species richness but also by
432 changing bryophyte species composition, whereas management-compatible species replace
433 the incompatible ones (Vellak and Ingerpuu, 2005).
434
435 4.3 Woodland indicator bryophyte species
436 Although coniferous plantations provide higher bryophyte species richness than deciduous
437 ones in our research areas, an increasing proportion of conifers had no benefiting effects on
438 woodland indicator bryophyte species. On the one hand, this might be biased by the
439 categorization of these species, as woodland indicator bryophyte species are largely linked to
440 ancient deciduous stands, whereas coniferous stands on ancient forest sites were less regarded
441 in studies on woodland indicator species (Preussing et al., 2011; Schmidt et al., 2011; Mölder
442 et al., 2015). On the other hand, our findings at least indicate the general preference of these
443 woodland species for deciduous forests, which are typical in our research areas, and implies
444 the failure of conifer plantations to provide suitable adequate refuges or alternative habitats
445 for them (Gustafsson et al., 1992). Confirming this, we found positive relations between litter
446 cover and negative relations to management intensity, both attributes of rather old, closed
447 deciduous forests, with woodland specialist bryophytes. However, conifer forests provide at
448 least habitat for some woodland indicator bryophyte species, especially deadwood specialists
449 (see also Coote et al., 2012).
450
451 4.4 Relation of herb and bryophyte species richness
452 The overall positive correlation between the richness of terricolous bryophyte and vascular
453 plant species reflects similar demands on light, moisture and nutrient conditions on the forest
454 floor (Tinya et al., 2009). Patterns of bryophyte species richness mainly resembled the ones of
455 vascular plants with respect to management impacts as observed in Boch et al. (2013b), as
456 well as Ingerpuu et al. (2003). However, the strength of correlation was lower than in earlier
457 studies (Pharo et al., 1999; McMullan-Fisher et al., 2010), and surprisingly even lower when
458 numbers of woodland specialists of bryophytes and vascular plants were compared. This
459 might be attributed to differing demands on soil properties as stated by Lalanne et al. (2008),
460 and adaptations of typical woodland vascular plant species (especially spring geophytes),
461 enabling their growth on the forest floor despite of large litter amounts contrary to specialized
462 woodland bryophytes (Cavard et al., 2011; Bartels and Chen, 2012), whose occurrence relies
463 mostly on availability of other substrates than bare soil (deadwood, bark, rocks) in dense
464 deciduous forests. Therefore, beech age-class forests might be suitable for typical woodland
465 vascular plants but not necessarily for bryophytes. This corresponds to the findings of Tullus
466 et al. (2018) who found destructive effects of shelterwood logging, resulting in age-class
467 forests, on bryophytes of conservation concern. Overall, our findings might imply that
468 woodland specialist bryophytes are better indicators for forest integrity than woodland
469 specialist vascular plants because bryophyte occurrence integrates substrate availability in
470 forests in addition to prevailing light, nutrient and humidity conditions.
471
472 5. Conclusions
473 Long-term effects of past management still affect current forest structure and bryophyte
474 communities in unmanaged forests, which highlight the importance of long-term
475 improvement of forest conditions ensuring the re-colonization of slow dispersing woodland
476 bryophyte species from refuges. Most important for bryophyte species richness is the
477 permanent availability of suitable substrates, which is not stringently linked to the
478 management type. Therefore managed age-class forests and selection forests can provide
479 important microhabitats for many bryophytes, including woodland indicator species, and
480 exceed species numbers of unmanaged forests. Due to the reliability of typical woodland
481 bryophytes as indicators of forest integrity, nature conservation efforts should focus on the
482 number and promotion on the total richness of this group. This includes the reduction of
483 management intensity, as well as maintaining and increasing habitat diversity and a variability
484 of substrates, such as coarse woody debris, the improvement of complex forest structures and
485 habitat heterogeneity by retaining patches with groups of old, mature to over-mature trees in
486 managed forests, maintaining forest climate conditions by silvicultural methods that assure
487 stand continuity, e.g. by selection cutting rather than clear cutting and shelterwood logging
488 (Humphrey et al., 2002; Gustafsson et al., 2004; Frego, 2007; Schulze et al., 2016).
489
490 6. Acknowledgements
491 We thank S. Gockel, A. Hemp, K. Wells, and S. Pfeiffer for maintaining the plot and project
492 infrastructure, and the late E.K.V. Kalko, K.E. Linsenmair, J. Nieschulze, I. Schöning, F.
493 Buscot, and W.W. Weisser for their role in setting up the Biodiversity Exploratories project.
494 The work has been funded by the DFG Priority Program 1374 "Infrastructure ‐ Biodiversity
495 Exploratories" (Fi-1246/6-1). Field work permits were issued by the responsible state
496 environmental offices of Baden-Württemberg, Thüringen, and Brandenburg.
497
498 7. References
499 Allan, E., Bossdorf, O., Dormann, C.F., Prati, D., Gossner, M.M., Tscharntke, T., Blüthgen,
500 N., Bellach, M., Birkhofer, K., Boch, S., Böhm, S., Börschig, C., Chatzinotas, A., Christ, S.,
501 Daniel, R., Diekötter, T., Fischer, C., Friedl, T., Glaser, K., Hallmann, C., Hodac, L., Hölzel,
502 N., Jung, K., Klein, A. M., Klaus, V.H., Kleinebecker, T., Krauss, J., Lange, M., Morris,
503 K.E., Müller, J., Nacke, H., Pasalic, E., Rillig, M.C., Rothenwöhrer, C., Schall, P., Scherber,
504 C., Schulze, W., Socher, S.A., Steckel, J., Steffan-Dewenter, I., Türke, M., Weiner, C.N.,
505 Werner, M., Westphal, C., Wolters, V., Wubet, T., Gockel, S., Gorke, M., Hemp, A.,
506 Renner, S.C., Schöning, I., Pfeiffer, S., König-Ries, B., Buscot, F., Linsenmair, K.E.,
507 Schulze, E.-D., Weisser, W.W., Fischer, M., 2014. Interannual variation in land-use
508 intensity enhances grassland multidiversity. P. Natl. Acad. Sci. USA 111, 308–313.
509 Ammer, C., Fichtner, A., Fischer, A., Gossner, M.M., Meyer, P., Seidl, R., Thomas, F.M.,
510 Annighöfer, P., Kreyling, J., Ohse, B., Berger, U., Feldmann, E., Häberle, K.-H., Heer, K.,
511 Heinrichs, S., Huth, F., Krämer-Klement, K., Mölder, A., Müller, J., Mund, M., Opgenoorth,
512 L., Schall, P., Scherer-Lorenzen, M., Seidel, D., Vogt, J., Wagner, S., in press. Key
513 ecological research questions for Central European forests. Basic Appl. Ecol.
514 https://doi.org/10.1016/j.baae.2018.07.006.
515 Andersson, L.I., Hytteborn, H., 1991. Bryophytes and decaying wood – a comparison
516 between managed and natural forest. Holarct. Ecol. 14, 121–30.
517 Aude, E., Lawesson, J.E., 1998. Vegetation in Danish beech forest: the importance of soil,
518 microclimate and management factors, evaluated by variation partitioning. Plant Ecol. 134,
519 53–65.
520 Bardat, J., Aubert, M., 2007. Impact of forest management on the diversity of corticolous
521 bryophyte assemblages in temperate forests. Biol. Conserv. 139, 47–66.
522 Barkman, J.J., 1958. Phytosociology and ecology of cryptogamic epiphytes. Van Gorcum,
523 Assen.
524 Bartels, S.F., Chen, H.Y.H., 2013. Interactions between overstorey and understorey
525 vegetation along an overstorey compositional gradient. J. Veg. Sci. 24, 415–588.
526 Bates, J.W., Farmer, A.M., (editors), 1992. Bryophytes and lichens in a changing
527 environment. Clarendon Press, Oxford.
528 Bates, J.W., Roy, D.B., Preston, C.D., 2004. Occurrence of epiphytic bryophytes in a
529 ‘‘tetrad’’ transect across southern Britain. 2. Analysis and modelling of epiphyte-
530 environmental relationships. J. Bryol. 26, 181–197.
531 Bauhus, J., Puettmann, K., & Messier, C., 2009. Silviculture forold-growth attributes. For.
532 Ecol. Manage. 258, 525–537.
533 BMELV (Bundesministerium für Ernährung, Landwirtschaft und Verbraucherschutz), 2012.
534 Bundeswaldinventur.
535 Boch, S., Berlinger, M., Fischer, M., Knop, E., Nentwig, W., Türke, M., Prati, D., 2013c.
536 Fern and bryophyte endozoochory by slugs. Oecologia 172, 817–822.
537 Boch, S., Müller, J., Prati, D., Blaser, S., Fischer, M., 2013d. Up in the tree - the overlooked
538 richness of bryophytes and lichens in tree crowns. PlosOne 8(12), e84913.
539 Boch, S., Prati, D., Hessenmöller, D., Schulze, E.-D., Fischer, M., 2013a. Richness of lichen
540 species, especially of threatened ones, is promoted by management methods furthering stand
541 continuity. PLoS ONE 8(1), e55461.
542 Boch, S., Prati, D., Müller, J., Socher, S.A., Baumbach, H., Buscot, F., Gockel, S., Hailer, J.,
543 Heinze, E., Hemp, A., Hessenmöller, D., Kalko, E.K.V. ,Linsenmair, K.E., Nieschulze, J.,
544 Pfeiffer, S., Pommer, U., Schöning, I., Schulze, E.-D., Seilwinder, C., Weisser, W.W.,
545 Wells,K., Fischer, M. 2013b. High plant species richness indicates management-related
546 disturbances rather than the conservation status of forests. Basic Appl. Ecol. 14, 496–505.
547 Borrass, L., Kleinschmit, D., Winkel, G. 2017. The “German model” of integrative
548 multifunctional forest management – Analysing the emergence and political evolution of a
549 forest management concept. For. Policy Econ. 77, 16–23.
550 Cavard, X., Bergeron, Y., Chen, H.Y.H., Pare, D., 2011. Effect of forest canopy composition
551 on soil nutrients and dynamics of the understorey: mixed canopies serve neither vascular nor
552 bryophyte strata. J. Veg. Sci. 22, 1105–1119.
553 Coote, L., French, L.J., Moore, K.M., Mitchell, F.J.G., Kelly, D.L., 2012. Can plantation
554 forests support plant species and communities of semi-natural woodland? For. Ecol.
555 Manage. 283, 86–95.
556 Davies, L., Bates, J.W., Bell, J.N.B., James, P.W., Purvis, O.W., 2007. Diversity and
557 sensitivty of epiphytes to oxides of nitrogen in London. Environ. Pollut. 146, 299–310.
558 Delgado, V., Ederra, A., 2013. Long-term changes (1982–2010) in the biodiversity of Spanish
559 beech forests assessed by means of Ellenberg indicator values of temperature, nitrogen, light
560 and pH. Biol. Conserv. 157, 99–107.
561 Dzwonko, Z., Gawroński, S., 2002. Effect of litter removal on species richness and
562 acidification of a mixed oak-pine woodland. Biol. Conserv. 106, 389–398.
563 Ewald, J., Jehl, H., Braun, L., Lohberger, E., 2011. Die Vegetation des Nationalparks
564 Bayerischer Wald als Ausdruck von Standort und Walddynamik. Tuexenia 31, 9–38.
565 Fengel, D., Wegener, G. 1984. Wood: chemistry, ultrastructure, reactions. Walter de Gruyter,
566 Berlin.
567 Fenton, N.J., Frego, K.A., Sims, M.R., 2003. Changes in forest floor bryophyte (moss and
568 liverwort) communities 4 years after forest harvest. Can. J. Bot. 81, 714–731.
569 Fischer, M., Bossdorf, O., Gockel, S., Hänsel, F., Hemp, A., Hessenmöller, D., Korte, G.,
570 Nieschulze, J., Pfeiffer, S., Prati, D., Renner, S., Schöning, I., Schumacher, U., Wells, K.,
571 Kalko, E.K.V., Buscot, F., Linsenmair, K. E., Schulze, E.-D., Weisser, W.W., 2010.
572 Implementing large-scale and long-term functional biodiversity research: the Biodiversity
573 Exploratories. Basic Appl. Ecol. 11, 473-485.
574 Frahm, J.-P., 2003. Climatic habitat differences of epiphytic lichens and bryophytes.
575 Cryptogam. Bryol. 24, 3–14.
576 Frego, K.A., 2007. Bryophytes as potential indicators of forest integrity. For. Ecol. Manage.
577 242, 65–75.
578 Friedel, A., Oheimb, G.V., Dengler, J., Härdtle, W., 2006. Species diversity and species
579 composition of epiphytic bryophytes and lichens – a comparison of managed and
580 unmanaged beech forests in NE Germany. Feddes Repert. 117, 172–185.
581 Friedel, A., Müller, F., 2004. Bryophytes and lichens as indicators for changes of air pollution
582 in the Serrahn Natural Forest Reserve (Mueritz National Park). Herzogia 17, 279–286.
583 Frisvoll, A., Prestø, T., 1997. Spruce forest bryophytes in central Norway and their
584 relationship to environmental factors including modern forestry. Ecography 20, 3–18.
585 Fritz, Ö., Gustafsson, L., Larsson, K., 2008. Does forest continuity matter in conservation? –
586 A study of epiphytic lichens and bryophytes in beech forests of southern Sweden. Biol.
587 Conserv. 141, 655–668.
588 Fritz, Ö., Caldiz, M.S., Brunet, J., 2009a. Interacting effects of tree characteristics on the
589 occurrence of rare epiphytes in a Swedish beech forest area. Bryologist 112, 488–505.
590 Fritz, Ö., Niklasson, M., Churski, M., 2009b. Tree age is a key factor for the conservation of
591 epiphytic lichens and bryophytes in beech forests. Appl. Veg. Sci. 12, 93–106.
592 Gerson, U., 1982. Bryophytes and invertebrates. In: Smith, A. J. E. (ed.). Bryophyte ecology.
593 Chapman & Hall, New York.
594 Gustafsson, L., Appelgren, L., Jonsson, F., Nordin, U., Persson, A., Weslien, J.-O., 2004.
595 High occurrence of red-listed bryophytes and lichens in mature managed forests in boreal
596 Sweden. Basic Appl. Ecol. 5, 123–129.
597 Gustafsson, L., Fiskesjö, A., Hallingback, T., Ingelog, T., Pettersson, B., 1992. Seminatural
598 deciduous broadleaved woods in southern Sweden – Habitat factors of importance to some
599 bryophytes. Biol. Conserv. 59, 175–181.
600 Hazell, P., Kellner, O., Rydin, H., Gustafsson, L., 1998. Presence and abundance of four
601 epiphytic bryophytes in relation to density of aspen (Populus tremula) and other stand
602 characteristics. For. Ecol. Manage. 107, 147–158.
603 Horvat, V., Heras, P., García-Mijangos, I., Biurrun, I., 2017. Intensive forest management
604 affects bryophyte diversity in the western Pyrenean silver fir-beech forests. Biol. Conserv.
605 215, 81–91.
606 Humphrey, J.W., Davey, S., Peace, A.J., Ferris, R., Harding, K., 2002. Lichens and bryophyte
607 communities of planted and seminatural forests in Britain: the influence of site type, stand
608 structure and deadwood. Biol. Conserv. 107, 165–180.
609 Huss, J., Butler-Manning, D., 2006. Entwicklungsdynamik eines buchendominierten
610 “Naturwald”-Dauerbeobachtungsbestandes auf Kalk im Nationalpark Hainich-Thüringen.
611 Waldökologie online 3, 59–73.
612 Ingerpuu, N., Vellak, K., Liira, J., Pärtel, M., 2003. Relationships between species richness
613 patterns in deciduous forests at the north Estonian limestone escarpment. J. Veg. Sci. 14,
614 773–780.
615 Kahl, T., Arnstadt, T., Baber, K., Bässler, C., Bauhus, J., Borken, W., Buscot, F., Floren, A.,
616 heibl, C., Hessenmöller, D., Hofrichter, M., Hoppe, B., Kellner, H., Krüger, D., Linsenmair,
617 K.E., Matzner, E., Otto, P., Purahong, W., Seilwinder, C., Schulze, E.D., Wende, B.,
618 Weisser, W.W., Gossner, M.M., 2017. Wood decay rates of 13 temeprate tree species in
619 relation to wood properties, enzyme activities and organismic diversity. For. Ecol. Manage.
620 391, 86–95.
621 Kaufmann, S., Hauck, M., Leuschner, C., 2017. Comparing the plant diversity of paired beech
622 primeval and production forests: Management reduces cryptogam, but not vascular plant
623 species richness. For. Ecol. Manage. 400, 58–67.
624 Kiebacher, T., Keller, C., Scheidegger, C., Bergamini, A., 2016. Hidden crown jewels: the
625 role of tree crowns for bryophyte and lichen species richness in sycamore maple wooded
626 pastures. Biodivers. Conserv. 25, 1605–1624.
627 Király, I., Ódor, P., 2010. The effect of stand structure and tree species composition on
628 epiphytic bryophytes in mixed deciduous–coniferous forests of Western Hungary. Biol.
629 Conserv. 143, 2063–2069.
630 Kriebitzsch, W.-U., Bültmann, H., Oheimb, G. v., Schmidt, M., Thiel, H., Ewald, J., 2013.
631 Forest-specific diversity of vascular plants, bryophytes, and lichens, in: Kraus, D., Krumm,
632 F. (Eds.), Integrative Approaches as an Opportunity for the Conservation of Forest
633 Biodiversity. European Forest Institute, Joensuu, pp. 158–169.
634 Lalanne, A., Bardat, J., Lalanne-Amara, F., Gautrot, T., Ponge, J.F., 2008. Opposite responses
635 of vascular plant and moss communities to changes in humus form, as expressed by the
636 Humus Index. J. Veg. Sci. 19, 645–652.
637 Manning, P., Gossner, M.M., Bossdorf, O., Allan, E., Zhang, Y.-Y., Prati, D., Blüthgen, N.,
638 Boch, S., Böhm, S., Börschig, C., Hölzel, N., Jung, K., Klaus, V.H., Klein, A.M.,
639 Kleinebecker, T., Krauss, J., Lange, M., Müller, J., Pašalić, E., Socher, S.A., Tschapka, M.,
640 Türke, M., Weiner, C., Werner, M., Gockel, S., Hemp, A., Renner, S.C., Wells, K., Buscot,
641 F., Kalko, E.K.V., Linsenmair, K.E., Weisser, W.W., Fischer, M., 2015. Grassland
642 management intensification weakens the associations among the diversities of multiple plant
643 and animal taxa. Ecology 96, 1492–1501.
644 Márialigeti, S., Németh B., Tinya, F., Ódor, P., 2009. The effects of stand structure on
645 ground-floor bryophyte assemblages in temperate mixed forests. Biodivers. Conserv. 18,
646 2223–2241.
647 McAlister, S., 1995. Species interactions and substrate specificity among log-inhabiting
648 bryophyte species. Ecology 76, 2184–2195.
649 McGee, G.G., Kimmerer, R.W., 2002. Forest age and management effects on epiphytic
650 bryophyte communities in Adirondack northern hardwood forests, New York, U.S.A. Can.
651 J. For. Res. 32, 1562–1576.
652 McMullan-Fisher, S.J.M., Kirkpatrick, J.B., May, T.W., Pharo, E.J., 2010. Surrogates for
653 macrofungi and mosses in reservation planning. Conserv. Biol. 24, 730–736.
654 Mežaka, A., Znotiņa, V., 2006. Epiphytic bryophytes in old growth forests of slopes, screes
655 and ravines in north-west Latvia. Acta Universitatis Latviensis 710, 103–116.
656 Mills, S.E., Macdonald, S.E., 2005. Factors influencing bryophyte assemblage at different
657 scales in the western Canadian boreal forest. Bryologist 108, 86–100.
658 Mitchell, R.J., Truscot, A.M., Leith, I.D., Cape, J.N., van Dijk, N., Tang, Y.S., Fowler, D.,
659 Sutton, M.A., 2005. A study of the epiphytic communities of Atlantic oak woods along an
660 atmospheric nitrogen deposition gradient. J. Ecol. 93, 482–292.
661 Mölder, A., Schmidt, M., Schönfelder, E., Engel, F., Schulz, F., 2015. Bryophytes as
662 indicators of ancient woodlands in Schleswig-Holstein (Northern Germany). Ecol. Indic. 54,
663 12–30.
664 Müller, J., Boch, S., Blaser, S., Fischer, M., Prati, D., 2015. Effects of forest management on
665 bryophyte communities on deadwood. Nova Hedwigia 100, 423–438.
666 Nelson, C.R., Halpern, C.B., 2005. Short-term effects of timber harvest and forest edges on
667 ground-layer mosses and liverworts. Can. J. Bot. 83, 610–620.
668 Nordén, B., Paltto, H., Götmark, F., Wallin, K., 2007. Indicators of biodiversity, what do they
669 indicate? – Lessons for conservation of cryptogams in oak-rich forest. Biol. Conserv. 135,
670 369–379.
671 Ódor, P., Heilmann-Clausen, J., Christensen, M., Aude, E., van Dort, K.W., Piltaver, A.,
672 Siller, I., Veerkamp, M.T., Standovár, T., van Hees, A.F.M., Kosec, J., Matocec, N.,
673 Kraigher, H., Grebenc, T., 2006. Diversity of dead wood inhabiting fungi and bryophytes in
674 semi-natural beech forests in Europe. Biol. Conserv. 131, 58–71.
675 Paillet, Y., Bergès, L., Hjältén, J., Ódor, P., Avon, C., Bernhardt-Römermann, M., Bijlsma,
676 R.-J., De Bruyn, L., Fuhr, M., Grandin, U., Kanka, R., Lundin, L., Luque, S., Magura, T.,
677 Matesanz, S., Mészáros, I., Sebastià, M.-T., Schmidt, W., Standovár, T., Tóthmérész, B.,
678 Uotila, A., Valladares, F., Vellak, K., Virtanen, R., 2010. Biodiversity differences between
679 managed and unmanaged forests: meta-analysis of species richness in Europe. Conserv.
680 Biol. 24, 101–112
681 Pharo, E.J., Beattie, A.J., Binns, D., 1999. Vascular plant diversity as a surrogate for
682 bryophyte and lichen diversity. Conserv. Biol. 13, 282–292.
683 Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D. and R Core Team, 2017. nlme: Linear and
684 Nonlinear Mixed Effects Models_. R package version 3.1–131, <URL: https://CRAN.R-
685 project.org/package=nlme>, (accessed 25 September 2018).
686 Preussing, M., Drehwald, U., Koperski, M., Thiel, M., Waesch, G., Baumann, M., Berg, C.,
687 Dierschke, H., Dolnik, C., Dürhammer, O., Ewald, J., Fischer, A., Grünberg, H., Heinken,
688 T., Jansen, F., Kison, H.-U., Klawitter, J., Kriebitzsch, W.-U., Loos, G.-H., Manthey, M.,
689 Müller, J., Paul, A., Reimann, M., Schmidt, M., Schmidt, W., Stetzka, K.M., Teuber, D.,
690 Teuber, U., Wagner, A., Wagner, I., Weckesser, M., Winter, S., Wolf, T., Wulf, M., 2011.
691 Waldartenliste der Moose Deutschlands. BfN-Skripten 299, 75–88.
